{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T17:33:21Z","timestamp":1784309601307,"version":"3.55.0"},"reference-count":56,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T00:00:00Z","timestamp":1751414400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T00:00:00Z","timestamp":1751414400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Optum Labs"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Electronic health records (EHR) contain data from disparate sources, spanning various biological and temporal scales. In this work, we introduce the Multiple Embedding Model for EHR (MEME), a deep learning framework for clinical decision support that operates over heterogeneous EHR. MEME first converts tabular EHR into \u201cpseudo-notes\u201d, reducing the need for concept harmonization across EHR systems and allowing the use of any state-of-the-art, open source language foundation models. The model separately embeds EHR domains, then uses a self-attention mechanism to learn the contextual importance of these multiple embeddings. In a study of 400,019 emergency department visits, MEME successfully predicted emergency department disposition, discharge location, intensive care requirement, and mortality. It outperformed traditional machine learning models (Logistic Regression, Random Forest, XGBoost, MLP), EHR foundation models (EHR-shot, MC-BEC, MSEM), and GPT-4 prompting strategies. Due to text serialization, MEME also exhibited strong few-shot learning performance in an external, unstandardized EHR database.<\/jats:p>","DOI":"10.1038\/s41746-025-01777-x","type":"journal-article","created":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T05:42:38Z","timestamp":1751434958000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Clinical decision support using pseudo-notes from multiple streams of EHR data"],"prefix":"10.1038","volume":"8","author":[{"given":"Simon A.","family":"Lee","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sujay","family":"Jain","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alex","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kyoka","family":"Ono","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arabdha","family":"Biswas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u00c1kos","family":"Rudas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jennifer","family":"Fang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeffrey N.","family":"Chiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,7,2]]},"reference":[{"key":"1777_CR1","doi-asserted-by":"publisher","first-page":"101693","DOI":"10.1016\/j.media.2020.101693","volume":"63","author":"N Tajbakhsh","year":"2020","unstructured":"Tajbakhsh, N. et al. Embracing imperfect datasets: a review of deep learning solutions for medical image segmentation. Med. Image Anal. 63, 101693 (2020).","journal-title":"Med. Image Anal."},{"key":"1777_CR2","doi-asserted-by":"publisher","first-page":"895","DOI":"10.1038\/s41573-023-00774-7","volume":"22","author":"MW Mullowney","year":"2023","unstructured":"Mullowney, M. W. et al. Artificial intelligence for natural product drug discovery. Nat. Rev. Drug Discov. 22, 895\u2013916 (2023).","journal-title":"Nat. Rev. Drug Discov."},{"key":"1777_CR3","doi-asserted-by":"publisher","first-page":"661","DOI":"10.1080\/17460441.2022.2090540","volume":"17","author":"A Farnoud","year":"2022","unstructured":"Farnoud, A., Ohnmacht, A. J., Meinel, M. & Menden, M. P. Can artificial intelligence accelerate preclinical drug discovery and precision medicine? Expert Opin. Drug Discov. 17, 661\u2013665 (2022).","journal-title":"Expert Opin. Drug Discov."},{"key":"1777_CR4","doi-asserted-by":"publisher","first-page":"107814","DOI":"10.1016\/j.compbiomed.2023.107814","volume":"169","author":"JL \u00c1vila-Jim\u00e9nez","year":"2024","unstructured":"\u00c1vila-Jim\u00e9nez, J. L., Cant\u00f3n-Habas, V., Carrera-Gonz\u00e1lez, M. D. P., Rich-Ruiz, M. & Ventura, S. A deep learning model for Alzheimer\u2019s disease diagnosis based on patient clinical records. Comput. Biol. Med. 169, 107814 (2024).","journal-title":"Comput. Biol. Med."},{"key":"1777_CR5","doi-asserted-by":"publisher","first-page":"e1083","DOI":"10.1016\/j.wneu.2024.05.052","volume":"187","author":"RP Kumar","year":"2024","unstructured":"Kumar, R. P. et al. Can artificial intelligence mitigate missed diagnoses by generating differential diagnoses for neurosurgeons? World Neurosurg. 187, e1083\u2013e1088 (2024).","journal-title":"World Neurosurg."},{"key":"1777_CR6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41698-023-00492-8","volume":"8","author":"S Khalighi","year":"2024","unstructured":"Khalighi, S. et al. Artificial intelligence in neuro-oncology: advances and challenges in brain tumor diagnosis, prognosis, and precision treatment. Npj Precis. Oncol. 8, 1\u201312 (2024).","journal-title":"Npj Precis. Oncol."},{"key":"1777_CR7","doi-asserted-by":"publisher","first-page":"951","DOI":"10.3390\/jpm13060951","volume":"13","author":"A Al Kuwaiti","year":"2023","unstructured":"Al Kuwaiti, A. et al. A review of the role of artificial intelligence in healthcare. J. Pers. Med. 13, 951 (2023).","journal-title":"J. Pers. Med."},{"key":"1777_CR8","doi-asserted-by":"publisher","first-page":"409","DOI":"10.7861\/clinmed.2022-0325","volume":"23","author":"EAA Idowu","year":"2023","unstructured":"Idowu, E. A. A., Teo, J., Salih, S., Valverde, J. & Yeung, J. A. Streams, rivers and data lakes: an introduction to understanding modern electronic healthcare records. Clin. Med. 23, 409\u2013413 (2023).","journal-title":"Clin. Med."},{"key":"1777_CR9","doi-asserted-by":"publisher","unstructured":"Zhao, W. X. et al. A survey of large language models. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2303.18223 (2023).","DOI":"10.48550\/arXiv.2303.18223"},{"key":"1777_CR10","doi-asserted-by":"publisher","unstructured":"Li, L. et al. A scoping review of using Large Language Models (LLMs) to investigate Electronic Health Records (EHRs). Preprint at https:\/\/doi.org\/10.48550\/arXiv.2405.03066 (2024).","DOI":"10.48550\/arXiv.2405.03066"},{"key":"1777_CR11","unstructured":"Radford, A., Narasimhan, K. Improving language understanding by generative pre-training (2018)."},{"key":"1777_CR12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-024-01166-w","volume":"7","author":"LL Guo","year":"2024","unstructured":"Guo, L. L. et al. A multi-center study on the adaptability of a shared foundation model for electronic health records. Npj Digit. Med. 7, 1\u20139 (2024).","journal-title":"Npj Digit. Med."},{"key":"1777_CR13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-023-00879-8","volume":"6","author":"M Wornow","year":"2023","unstructured":"Wornow, M. et al. The shaky foundations of large language models and foundation models for electronic health records. Npj Digit. Med. 6, 1\u201310 (2023).","journal-title":"Npj Digit. Med."},{"key":"1777_CR14","doi-asserted-by":"crossref","unstructured":"Alsentzer, E. et al. Publicly available clinical BERT embeddings. Proceedings of the 2nd Clinical Natural Language Processing Workshop (2019).","DOI":"10.18653\/v1\/W19-1909"},{"key":"1777_CR15","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1038\/s41586-023-05881-4","volume":"616","author":"M Moor","year":"2023","unstructured":"Moor, M. et al. Foundation models for generalist medical artificial intelligence. Nature 616, 259\u2013265 (2023).","journal-title":"Nature"},{"key":"1777_CR16","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-62922-y","volume":"10","author":"Y Li","year":"2020","unstructured":"Li, Y. et al. BEHRT: transformer for electronic health records. Sci. Rep. 10, 7155 (2020).","journal-title":"Sci. Rep."},{"key":"1777_CR17","doi-asserted-by":"publisher","unstructured":"Rupp, M., Peter, O. & Pattipaka, T. ExBEHRT: extended transformer for electronic health records. in Trustworthy Machine Learning for Healthcare (eds. Chen, H. & Luo, L.) 73\u201384. https:\/\/doi.org\/10.1007\/978-3-031-39539-0_7 (Springer Nature, 2023).","DOI":"10.1007\/978-3-031-39539-0_7"},{"key":"1777_CR18","doi-asserted-by":"publisher","first-page":"1110","DOI":"10.13063\/2327-9214.1110","volume":"2","author":"R Makadia","year":"2014","unstructured":"Makadia, R. & Ryan, P. B. Transforming the Premier Perspective\u00ae Hospital Database into the Observational Medical Outcomes Partnership (OMOP) Common Data Model. eGEMs 2, 1110 (2014).","journal-title":"eGEMs"},{"key":"1777_CR19","unstructured":"Arnrich, B. et al. Medical event data standard (MEDS): Facilitating machine learning for health. Workshop on Learning from Time Series For Health (ICLR, 2024)."},{"key":"1777_CR20","first-page":"24322","volume":"36","author":"M McDermott","year":"2023","unstructured":"McDermott, M., Nestor, B., Argaw, P. & Kohane, I. S. Event Stream GPT: a data pre-processing and modeling library for generative, pre-trained transformers over continuous-time sequences of complex events. Adv. Neural Inf. Process. Syst. 36, 24322\u201324334 (2023).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"1777_CR21","doi-asserted-by":"publisher","first-page":"103637","DOI":"10.1016\/j.jbi.2020.103637","volume":"113","author":"E Steinberg","year":"2021","unstructured":"Steinberg, E. et al. Language models are an effective representation learning technique for electronic health record data. J. Biomed. Inform. 113, 103637 (2021).","journal-title":"J. Biomed. Inform."},{"key":"1777_CR22","first-page":"45794","volume":"36","author":"E Chen","year":"2023","unstructured":"Chen, E. et al. Multimodal Clinical Benchmark for Emergency Care (MC-BEC): a comprehensive benchmark for evaluating foundation models in emergency medicine. Adv. Neural Inf. Process. Syst. 36, 45794\u201345811 (2023).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"1777_CR23","doi-asserted-by":"crossref","unstructured":"Yan, An. et al. RadBERT: adapting transformer-based language models to radiology. Radiol. Artif. Intell. 4.4, e210258 (2022).","DOI":"10.1148\/ryai.210258"},{"key":"1777_CR24","first-page":"67125","volume":"36","author":"M Wornow","year":"2023","unstructured":"Wornow, M., Thapa, R., Steinberg, E., Fries, J. & Shah, N. EHRSHOT: an EHR benchmark for few-shot evaluation of foundation models. Adv. Neural Inf. Process. Syst. 36, 67125\u201367137 (2023).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"1777_CR25","doi-asserted-by":"publisher","first-page":"103637","DOI":"10.1016\/j.jbi.2020.103637","volume":"113","author":"Ethan Steinberg","year":"2021","unstructured":"Steinberg, Ethan et al. Language models are an effective representation learning technique for electronic health record data. J. Biomed. Inform. 113, 103637 (2021).","journal-title":"J. Biomed. Inform."},{"key":"1777_CR26","doi-asserted-by":"publisher","unstructured":"Li, Y., Wehbe, R. M., Ahmad, F. S., Wang, H. & Luo, Y. Clinical-Longformer and Clinical-BigBird: Transformers for long clinical sequences. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2201.11838 (2022).","DOI":"10.48550\/arXiv.2201.11838"},{"key":"1777_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-022-01782-9","volume":"9","author":"F Xie","year":"2022","unstructured":"Xie, F. et al. Benchmarking emergency department prediction models with machine learning and public electronic health records. Sci. Data 9, 658 (2022).","journal-title":"Sci. Data"},{"key":"1777_CR28","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-022-01899-x","volume":"10","author":"AEW Johnson","year":"2023","unstructured":"Johnson, A. E. W. et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci. Data 10, 1 (2023).","journal-title":"Sci. Data"},{"key":"1777_CR29","unstructured":"Hegselmann, S. et al. TabLLM: few-shot classification of tabular data with large language models. In: Proc. 26th International Conference on Artificial Intelligence and Statistics 5549\u20135581 (PMLR, 2023)."},{"key":"1777_CR30","unstructured":"Hegselmann, S. et al. A data-centric approach to generate faithful and high quality patient summaries with large language models. Conference on Health, Inference, and Learning (PMLR, 2024)."},{"key":"1777_CR31","unstructured":"Ellershaw, S. et al. Automated generation of hospital discharge summaries using clinical guidelines and large language models. In: Proc. 2024 Spring Symposium on Clinical Foundation Models at Stanford University, Stanford, California (AAAI, 2024)."},{"key":"1777_CR32","doi-asserted-by":"crossref","unstructured":"Yuan, D. et al. A continued pretrained LLM approach for automatic medical note generation. (NAACL (Short Papers), 2024).","DOI":"10.18653\/v1\/2024.naacl-short.47"},{"key":"1777_CR33","unstructured":"Textualization of Oral Epics. https:\/\/www.degruyter.com\/document\/doi\/10.1515\/9783110825848\/pdf?licenseType=restricted#page=11."},{"key":"1777_CR34","doi-asserted-by":"publisher","unstructured":"Vasantharajan, C. et al. MedBERT: a pre-trained language model for biomedical named entity recognition. In: Proc. Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) 1482\u20131488. https:\/\/doi.org\/10.23919\/APSIPAASC55919.2022.9980157 (IEEE, 2022).","DOI":"10.23919\/APSIPAASC55919.2022.9980157"},{"key":"1777_CR35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-021-00455-y","volume":"4","author":"L Rasmy","year":"2021","unstructured":"Rasmy, L., Xiang, Y., Xie, Z., Tao, C. & Zhi, D. Med-BERT: pretrained contextualized embeddings on large-scale structured electronic health records for disease prediction. Npj Digit. Med. 4, 1\u201313 (2021).","journal-title":"Npj Digit. Med."},{"key":"1777_CR36","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1109\/JBHI.2023.3327951","volume":"28","author":"K Hur","year":"2024","unstructured":"Hur, K. et al. GenHPF: general healthcare predictive framework for multi-task multi-source learning. IEEE J. Biomed. Health Inform. 28, 502\u2013513 (2024).","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"1777_CR37","doi-asserted-by":"publisher","first-page":"e0118432","DOI":"10.1371\/journal.pone.0118432","volume":"10","author":"T Saito","year":"2015","unstructured":"Saito, T. & Rehmsmeier, M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLOS ONE 10, e0118432 (2015).","journal-title":"PLOS ONE"},{"key":"1777_CR38","doi-asserted-by":"publisher","unstructured":"OpenAI et al. GPT-4 Technical Report. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2303.08774 (2024).","DOI":"10.48550\/arXiv.2303.08774"},{"key":"1777_CR39","unstructured":"Grattafiori, A. et al. The llama 3 herd of models. Preprint at: 10.48550\/arXiv.2407.21783 (2023)."},{"key":"1777_CR40","first-page":"22199","volume":"35","author":"T Kojima","year":"2022","unstructured":"Kojima, T. et al. Large language models are zero-shot reasoners. Adv. Neural Inf. Process. Syst. 35, 22199\u201322213 (2022).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"1777_CR41","doi-asserted-by":"publisher","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","volume":"36","author":"J Lee","year":"2020","unstructured":"Lee, J. et al. BioBERT: a pre-trained biomedical language representation model for biomedical text mining. Bioinformatics 36, 1234\u20131240 (2020).","journal-title":"Bioinformatics"},{"key":"1777_CR42","unstructured":"Anthropic. Prompt Engineering Overview. Anthropic Documentation, https:\/\/docs.anthropic.com\/en\/docs\/build-with-claude\/prompt-engineering\/overview."},{"key":"1777_CR43","doi-asserted-by":"crossref","unstructured":"Brown, K. E. et al. Large language models are less effective at clinical prediction tasks than locally trained machine learning models. J. Am. Med. Inform. Assoc. 32, 811\u2013822 (2025).","DOI":"10.1093\/jamia\/ocaf038"},{"key":"1777_CR44","doi-asserted-by":"crossref","unstructured":"Wang, Y., Yao, Q., Kwok, J. T. & Ni, L. M. Generalizing from a few examples: a survey on few-shot learning. ACM Comput. Surv. 53, 63:1\u201363:34 (2020).","DOI":"10.1145\/3386252"},{"key":"1777_CR45","doi-asserted-by":"publisher","unstructured":"Parnami, A. & Lee, M. Learning from few examples: a summary of approaches to few-shot learning. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2203.04291 (2022).","DOI":"10.48550\/arXiv.2203.04291"},{"key":"1777_CR46","doi-asserted-by":"publisher","unstructured":"Liu, J. et al. Towards out-of-distribution generalization: a survey. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2108.13624 (2023).","DOI":"10.48550\/arXiv.2108.13624"},{"key":"1777_CR47","unstructured":"Hill, B. L. et al. CHIRon: a generative foundation model for structured sequential medical data. In: Proc. NeurIPS 2023 - Generative AI for Science Workshop (2023)."},{"key":"1777_CR48","unstructured":"Spackman, K. A., Campbell, K. E. & C\u00f4t\u00e9, R. A. SNOMED RT: a reference terminology for health care. In: Proc. AMIA Annual Fall Symposium 640\u2013644 (AMIA, 1997)."},{"key":"1777_CR49","first-page":"11763","volume":"35","author":"T Dinh","year":"2022","unstructured":"Dinh, T. et al. LIFT: language-interfaced fine-tuning for non-language machine learning tasks. Adv. Neural Inf. Process. Syst. 35, 11763\u201311784 (2022).","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"1777_CR50","unstructured":"Lee, S. A. & Ono, K. Text serialization and their relationship with the conventional paradigms of tabular machine learning. AI for Science Workshop (ICML, 2024)."},{"key":"1777_CR51","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-024-01127-3","volume":"7","author":"L Goetz","year":"2024","unstructured":"Goetz, L., Seedat, N., Vandersluis, R. & van der Schaar, M. Generalization\u2014a key challenge for responsible AI in patient-facing clinical applications. Npj Digit. Med. 7, 1\u20134 (2024).","journal-title":"Npj Digit. Med."},{"key":"1777_CR52","doi-asserted-by":"publisher","unstructured":"Bommasani, R. et al. On the opportunities and risks of foundation models. Preprint at https:\/\/doi.org\/10.48550\/arXiv.2108.07258 (2022).","DOI":"10.48550\/arXiv.2108.07258"},{"key":"1777_CR53","doi-asserted-by":"crossref","unstructured":"Hong, W. S. et al. Association of red blood cell distribution width with hospital admission and in-hospital mortality across all-cause adult emergency department visits. JAMIA Open 6, ooad053 (2023).","DOI":"10.1093\/jamiaopen\/ooad053"},{"key":"1777_CR54","doi-asserted-by":"publisher","unstructured":"Grootendorst, M. BERTopic: neural topic modeling with a class-based TF-IDF procedure. https:\/\/doi.org\/10.48550\/arXiv.2203.05794 (2022).","DOI":"10.48550\/arXiv.2203.05794"},{"key":"1777_CR55","doi-asserted-by":"publisher","first-page":"100243","DOI":"10.1016\/j.xgen.2022.100243","volume":"3","author":"R Johnson","year":"2023","unstructured":"Johnson, R. et al. The UCLA ATLAS Community Health Initiative: promoting precision health research in a diverse biobank. Cell Genom. 3, 100243 (2023).","journal-title":"Cell Genom."},{"key":"1777_CR56","unstructured":"Vaswani, A. et al. Attention is all you need. Adv. Neural Inf. Process. Syst. 30 (2017)."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01777-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01777-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01777-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T05:42:49Z","timestamp":1751434969000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-025-01777-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,2]]},"references-count":56,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1777"],"URL":"https:\/\/doi.org\/10.1038\/s41746-025-01777-x","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,2]]},"assertion":[{"value":"1 October 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 June 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 July 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"J.C. receives research funding from Optum Labs. The other authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"394"}}